Staphylococcus aureus is the leading cause of bloodstream, lower respiratory tract, skin and soft tissue infections in the United States and antibiotic resistant strains (MRSA) are isolated in more than half of all community and hospital infections. The search for protective immunity against invasive S. aureus disease has been a premier research goal since the discovery of this microbe. While several envelope components and secreted products have been investigated as vaccine antigens (surface proteins, capsular polysaccharide, exopolysaccharide, a-hemolysin, and leukocidins), no single antigen has facilitated the development of long-term protective immunity. Furthermore, investigations to date have not yielded a clear understanding of the features of the host immune response that are requisite for the generation of such protective immunity. This proposal is based on three fundamental discoveries. 1. S. aureus modulates immune responses during infection and prevents the development of protective immunity. The recent analysis of staphylococcal variants in the ess and agr pathways that are capable of generating protective immunity in live vaccination models now enables the characterization of protective antigens via genetic subtraction. 2. Levels of protective immunity against staphylococcal infection in animal models are increased when two or more antigens are combined. The reciprocal phenomenon is also true: genetic subtraction of more than one vaccine antigen (virulence factor) leads to a commensurate reduction of staphylococcal virulence. 3. For infectious processes in the lung, secretion of a-hemolysin is an essential virulence attribute of staphylococci, enabling the design of unique vaccine and immunotherapeutic strategies. Based on these discoveries, we propose a rational approach for vaccine testing, whereby the inclusion of individual subunits must be based on three criteria, (i) Genes specifying vaccine antigens must be present in relevant clinical isolates of S. aureus. (ii) Staphylococcal mutants lacking these genes must reveal a contribution of these components to the pathogenesis of S. aureus infections, (iii) Immunization with individual subunit components or a combination of subunits must generate immune responses that improve the outcome of S. aureus infections when tested with multiple animal models of disease.